KoboToolbox for AI agents
KoboToolbox is a data collection platform used by researchers and humanitarian organizations to build forms and gather field survey data. Connect it once through Arc0, and your agent, or Claude, ChatGPT and Cursor, can use it through one MCP endpoint, limited to what each user approved.
What agents do in KoboToolbox.
Export survey submissions
Export the submissions collected under a deployed project for analysis in another tool.
Check a project's submission count
List recent submissions to a project to confirm field data collection is on track.
Confirm before deleting a project
Require approval before deleting a project, since it removes its forms and collected submissions.
9 KoboToolbox actions, graded by risk.
Every KoboToolbox action is tagged read, write or destructive, so one policy covers the whole app and new actions inherit the right default.
read
5Look things up. Allowed by default.
- kobotoolbox.get_projectGet Project
- kobotoolbox.list_assetsList Assets
- kobotoolbox.get_submissionGet Submission
- kobotoolbox.list_submissionsList Submissions
- kobotoolbox.export_submissionsExport Submissions
write
2Create and change things. Allow, or ask the user first.
- kobotoolbox.create_projectCreate Project
- kobotoolbox.deploy_projectDeploy Project
destructive
2Delete, cancel or archive. Ask first, or deny outright.
- kobotoolbox.delete_exportDelete Export
- kobotoolbox.delete_projectDelete Project
KoboToolbox in three steps.
- 01Your users connect KoboToolboxThey add their KoboToolbox api key on Arc0 Connect, under your brand. It goes straight into the vault.
- 02You set the rulesReads run, writes like “create Project” can wait for the user, and “delete Export” can be denied outright.
- 03Any agent can actYour agent calls KoboToolbox through the Arc0 SDK or MCP, and so can Claude, ChatGPT and Cursor. Every call lands on the audit log.
await arc0.policies.set('kobotoolbox', { read: 'allow', write: 'ask', // create_project destructive: 'deny', // delete_export }) # Claude Code: the same connection, one URL $ claude mcp add --transport http arc0 \ https://mcp.arc0.ai/u/u_8f2
How KoboToolbox connects.
Users add their KoboToolbox api key on Arc0 Connect. It is encrypted in the vault, never shown to the model, and each user can rotate or revoke it at any time.
The same KoboToolbox connection serves your agent over MCP and your own backend over REST and the proxy, so a user connects once. How Arc0 handles credentials →
- AUTH
- API key
- CREDENTIALS
- Per-tenant encrypted vault
- MODEL SEES
- Results only, never credentials
- AUDIT LOG
- Every call, on every plan
Use KoboToolbox from any agent.
ClaudeChatGPTCursor
Codex
VS CodeOpenAI Agents SDKClaude Agent SDK
Vercel AI SDK
Mastra
LangGraph
KoboToolbox and Arc0, answered.
Can I use KoboToolbox with Claude, ChatGPT or Cursor?
Yes. Connect KoboToolbox to Arc0 once, then add your Arc0 MCP URL to Claude, ChatGPT, Cursor, Claude Code or any other remote-MCP client. Each assistant only gets the KoboToolbox actions you allow.
How do users connect KoboToolbox?
Users add their KoboToolbox api key on Arc0 Connect. It is encrypted in the vault, never shown to the model, and each user can rotate or revoke it at any time.
Which KoboToolbox actions can my agent take?
9 in total: 5 read, 2 write and 2 destructive, such as “create Project”. Your policies decide which of them each agent may call.
Can I stop my agent from deleting things in KoboToolbox?
Yes. Actions like “delete Export” are graded destructive. Set destructive actions to deny, or to ask so the user approves each one, and blocked calls still show up on the audit log.
Can my own backend call KoboToolbox too?
Yes. The same KoboToolbox connection is available over REST and through the Arc0 proxy, so your product and your agent share one connection per user.
Plug KoboToolbox into your agent.
Your users connect KoboToolbox once, under your brand. Your agent gets 9 actions behind your policies, with every call on the record.
Free to build · MCP + REST · Audit log on every plan